Prevalence and Demographic Features of Childhood Growth Hormone

European Journal of Endocrinology (2004) 151 67–72 ISSN 0804-4643

CLINICAL STUDY Prevalence and demographic features of childhood growth hormone deﬁciency in Belgium during the period 1986–2001 M Thomas1, G Massa1,2, M Craen3, F de Zegher4, J P Bourguignon5, C Heinrichs2, J De Schepper6, M Du Caju7, G Thiry-Counson9 and M Maes8 1Belgian Study Group for Paediatric Endocrinology, and Departments of Paediatrics of the Universities of 2Bruxelles, 3Gent, 4Leuven, 5Lie`ge, 6Brussel, 7Antwerpen, 8Louvain, Belgium and 9Clinique de l’Espe´rance, Saint-Nicolas, Belgium (Correspondence should be addressed to Marc Maes, De´partement de Pe´diatrie, Cliniques Universitaires Saint-Luc, Avenue Hippocrate, 10, 1200 Bruxelles, Belgium; Email: [email protected])

Abstract Objective: Since the availability of recombinant human growth hormone (rhGH) all children with growth hormone deficiency (GHD) living in Belgium are offered rhGH treatment after approval by a peer–review board. In this study, we evaluated the prevalence and demographic features of childhood GHD in Belgium during the period 1986–2001 and we compared them with the data from other countries. Methods: Diagnostic, demographic and baseline auxological data of 714 children diagnosed as having GHD between 1986 and 2001 were retrieved from the database of the Belgian Study Group for Pae- diatric Endocrinology. Results: The prevalence of GHD in Belgium was estimated to be 1/5600. The origin of GHD was idiopathic (idGHD) in 41% of the patients, congenital (congGHD) in 20% and acquired (acqGHD) in 35%. During the first 4 years (1986–1989) more patients were classified as idGHD; thereafter the distribution between the three aetiology groups did not change. In all groups, boys outnumbered girls but this preponderance was especially pronounced in congGHD patients (male:female ¼ 4:1) with a central malformation that associates an anterior pituitary hypoplasia, a missing, fine or normal pituitary stalk and an ectopic posterior pituitary. Thirteen percent of the patients with idGHD, 50% with congGHD and 52% with acqGHD had multiple pituitary deficiencies. Patients with congGHD were the youngest (mean^S.D. age: 6.5^4.7 years) and were the shortest (23.0^1.3 standard deviation score (SDS)) at the start of rhGH treatment. There was no time trend over the studied period for age and height at onset of GH therapy. Conclusion: In Belgium, the prevalence of childhood GHD can be estimated as 1/5600 which is comparable to other recent surveys. The yearly number of new patients for the different aetiologies and the auxological parameters have remained relatively constant over the last 16 years.

European Journal of Endocrinology 151 67–72

Introduction Endocrinology (BSGPE) continued to peer-review the records of patient candidates for rhGH treatment Growth hormone deficiency (GHD) is a relatively before applying for its reimbursement. uncommon cause of growth retardation and short sta- In the present study, we evaluated the prevalence ture. The prevalence of childhood GHD reported in the and described demographic features of childhood GHD literature diverges widely varying between 1/3480 and in Belgium during the period 1986–2001, and we 1/30 000 children (1–8). compared the results with those from other databases In 1986, recombinant human growth hormone and other countries. The data in this report present a (rhGH) was licensed for use in Belgium and since then profile of what could be considered standard diagnostic supplies of rhGH have been virtually unlimited. practices in Belgium. Growth hormone deficiency remains the primary indi- cation for rhGH treatment and all children with GHD can be offered optimal treatment. Treatment with Patients and methods rhGH is reimbursed by the Belgian government, and until 1993, the use of rhGH was supervised by a minis- In Belgium, the files of children who are candidates terial commission of experts. After the dismissal of the for treatment with rhGH are discussed by the mem- ministerial commission, the paediatric endocrinologists bers of the BSGPE using a systematic peer-reviewed belonging to the Belgian Study Group for Paediatric process. The BSGPE includes the Belgian paediatric

q 2004 Society of the European Journal of Endocrinology Online version via http://www.eje.org

Downloaded from Bioscientifica.com at 10/01/2021 07:58:02AM via free access 68 M Thomas and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2004) 151 endocrinologists treating children with growth dis- was below 22 SDS. Mid-parental height SDS was calcu- orders, and represents the seven Belgian universities. lated as (father’s height SDS þ mother’s height SDS)/2. An application for treatment with rhGH to be reim- The prevalence of GHD was calculated on the basis of bursed by the government is only accepted after a the Belgian demographic data provided by the National consensus decision by the majority of the BSGPE Institute of Statistics (13). Results are expressed as members. Baseline and treatment auxological and bio- means^S.D. or median (range) as indicated. Compari- chemical data from the patients approved for rhGH sons between the three aetiology groups were carried treatment are stored in the database of the BSGPE. out with the chi-squared test, or by ANOVA or Signed informed consent forms are obtained from Kruskal–Wallis test. The Jonckheere–Terpstra non- the parents and the patients if older than 12 years. parametric test was used to test for time trends for From the database of the BSGPE, the data of all chil- age and height at onset of GH therapy. A P , 0.05 dren (n ¼ 740) diagnosed with GHD between 1986 value was considered significant. and 2001 were retrieved. The data of 26 patients were incomplete and they were not included in the analysis. GHD was defined by auxological and biological criteria: Results height velocity ,the 10th or ,the 25th centile for Prevalence of GHD patients with idiopathic and organic GHD respectively, and peak serum GH levels, measured in the individual For the studied period, 714 (442 boys) patients with institution’s laboratory, ,10 ng/ml during two stimu- documented GHD were retrieved from the database of lation tests, usually the glucagon stimulation test and the BSGPE. Yearly, between 29 and 66 (median 45) the insulin tolerance test (9). Sex steroid priming was new GHD patients started treatment. In 2001, 386 carried out for boys aged 10 years or older and for girls GHD patients between 0 and 18 years were treated aged 9 years or older if they had a pubertal development with rhGH. Consequently, for the Belgian population, # Genital stage 2 or Breast stage 2, independently of counting about 2.2 million children younger than 18 bone age. In boys, 50 mg testosterone propionate were years, a prevalence (95% confidence interval) of GHD administered intramuscularly 72 h before the GH stimu- children of 18 (16–20) per 100 000 (or 1/5600) was lation test, and the girls received orally 50 mg ethinyloes- estimated. tradiol during the three days before the test (9). GHD was considered as complete when the maximal GH values at Aetiology and gender the two tests were below 5 ng/ml; GHD was considered as partial when the maximal GH values at the two tests The aetiology of GHD and the gender of the patients are were below 10 ng/ml with at least one above 5 ng/ml given in Table 1. Forty-one per cent of the patients were (9). Cranial imaging studies, with special emphasis on classified as idGHD, 20% as congGHD and 35% as the hypothalamo–pituitary region, were performed in acqGHD; 4% of the patients had a defined syndrome all patients using computed tomography (CT) scanning with concomitant GHD. The aetiology of the acqGHD or magnetic resonance imaging (MRI). Before 1990, was due, for the majority of the patients (93%), to an MRI was performed in 29% of the patients with cranial tumours and/or cerebral radiotherapy. Boys non-acquired GHD, 1990–1993 MRI was performed in outnumbered girls in all groups. The preponderance 76% and 1994–1997 in 89% of these patients. After of boys was especially pronounced in the patients 1998, 98% of the patients were evaluated by MRI. with congGHD represented mainly by patients with a The data collected in this study included aetiology of central malformation, most frequently the association GHD, gender, age and height at the start of rhGH treat- of hypoplastic anterior pituitary, missing, fine or ment, birth weight, parental heights, and associated normal stalk and ectopic posterior pituitary (sex ratio pituitary deficiencies. The aetiologies of GHD were classi- male:female ¼ 4). fied according to the Kabi International Growth Study Figure 1 shows the 4-yearly evolution of new aetiology classification list (10). Patients were classified patients with GHD. During the first 4 years (1986– as idiopathic GHD (idGHD) when no anomaly could be 1989) more patients were classified as idGHD. The detected after the performed investigations. Patients yearly number of new patients with congGHD and were considered as congenital GHD (congGHD) when a acqGHD remained relatively stable over the whole cerebral developmental anomaly was documented or period. when a genetic cause of GHD was found. Patients with a postnatally acquired cerebral anomaly were classified Severity of GHD and associated pituitary as acquired GHD (acqGHD). Height was measured deficiencies with a Harpenden stadiometer and expressed as standard deviation score (SDS) using the Tanner 1966 refer- Of the patients with idGHD, 37% had complete GHD ence data (11). Birth weight SDS was calculated using while 78% of the patients with congGHD and 62% of the Swedish reference data (12). The patients were con- the patients with acqGHD were diagnosed as complete sidered as small for gestational age (SGA) if birth weight GHD (P , 0.001).

www.eje.org

Downloaded from Bioscientifica.com at 10/01/2021 07:58:02AM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2004) 151 Epidemiology of childhood GHD 69

Table 1 Aetiology of GHD. The numbers of patients in the congenital and the acquired GHD groups (both belonging to the organic GHD group) were further subdivided depending on their speciﬁc aetiology.

Number of patients Sex ratio M:F Group No. No. in ﬁrst No. in second in group subgroup subgroup % of total Group Subgroup

Idiopathic GHD 296 41 1.5 Organic GHD Congenital form 142 20 3.0 Genetic cause (Pit-1, GH1, HESX-1 defect) 3 Central malformation 136 HME1 101 4.0 Other2 35 1.9 Other3 3 Acquired GHD 245 35 1.2 Cranial tumour of the pituitary–hypothalamus area4 77 Irradiated 37 Non-irradiated 40 Cranial tumour distant from the 119 pituitary–hypothalamus area5 Irradiated 110 Non-irradiated 9 Treatment for malignancy outside the cranium 31 Acute lymphatic leukaemia 25 Lymphoma and other tumours outside the cranium6 6 Other causes of acquired GHD7 18 Syndromes with concomitant GHD 31 4 2.9 Total 714 1.6

1 Hypoplastic anterior pituitary–missing stalk, ﬁne stalk or normal stalk–ectopic posterior pituitary. 2 Septo-optic dysplasia (n ¼ 8), mid-line palatal cleft (n ¼ 5), arachnoid cyst (n ¼ 4), congenital hydrocephalus (n ¼ 9), empty sella syndrome (n ¼ 5), solitary central maxillary incisor syndrome (n ¼ 1), schizencephalia (n ¼ 1), encephaloclastic lesion (n ¼ 1), Rathke’s cyst (n ¼ 1). 3 Syndromes with congenital GHD: q18 deletion (n ¼ 1), 46,XX/46,XX del(20) (p122) (n ¼ 1); congenital toxoplasmosis (n ¼ 1). 4 Craniopharyngioma (n ¼ 56), germinoma (n ¼ 12), adenoma (n ¼ 5), chordoma (n ¼ 1), choriocarcinoma (n ¼ 1), hypothalamic tumour (n ¼ 2). 5 Medulloblastoma (n ¼ 47), astrocytoma (n ¼ 24), glioma (n ¼ 15), ependymoma (n ¼ 7), rhabdomyosarcoma (n ¼ 12), osteosarcoma (n ¼ 1), gangliocy- toma (n ¼ 2), glioblastoma (n ¼ 1), meningioma (n ¼ 1), mesencephalic tumour (n ¼ 1), neurocytoma (n ¼ 1), papilloma (n ¼ 1), primitive neuroectodermal turmour (n ¼ 1), retinoblastoma (n ¼ 2), teratoma (n ¼ 2), other (n ¼ 1). 6 Lymphoma (n ¼ 4), neuroblastoma/TBI (n ¼ 1), hepatoblastoma (n ¼ 1). 7 Head trauma (n ¼ 8), histiocytosis (n ¼ 6), enlarged stalk (n ¼ 1), CNS infection (n ¼ 1), Total body irradiation for thalassaemia major (n ¼ 1), Blackfan- Diamond (n ¼ 1).

Table 2 shows the associated pituitary deficiencies. adrenocorticotrophin (ACTH) deficiency more fre- Fifty per cent of the patients with congGHD and 52% quently than patients with idGHD (P , 0.01). In the with acqGHD had other hormonal deficiencies, while three groups, about half of the patients with multiple these were present only in 13% of the patients with pituitary deficiencies had luteinizing hormone/follicle- idGHD (P , 0.01). Thyrotrophin (TSH) deficiency was stimulating hormone (LH/FSH) deficiency. More than the most frequent pituitary deficiency associated with half of the patients with multiple acqGHD had an GHD. Patients with congGHD and acqGHD had an associated antidiuretic (ADH) deficiency.

Age and height at onset of therapy, birth weight and parental height Table 3 shows the age and height at onset of rhGH therapy, birth weight and parental height for the three aetiology groups. For the whole group of patients, rhGH therapy was started at the median age of 9.1 years (range 0.0–27.3 years). The patients with congGHD were younger than the patients in the other groups (P , 0.001). In the three groups, no differences were found for the age at onset of GH therapy between patients with partial and complete GHD. In the patients Figure 1 Four-yearly evolution of the number of new patients with with idGHD and congGHD rhGH treatment was started idGHD (black bars), congGHD (dark grey bars) and acqGHD when a serious height deﬁcit (22.8^0.9 SDS and (light grey bars). 23.0^1.3 SDS respectively) was already present.

www.eje.org

Downloaded from Bioscientifica.com at 10/01/2021 07:58:02AM via free access 70 M Thomas and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2004) 151

Table 2 Number of patients with multiple pituitary hormone deﬁciencies (MPHD) and distribution of other pituitary deﬁciencies for the different aetiologies.

MPHD TSH/MPHD ACTH/MPHD LH FSH/MPHD ADH/MPHD Cause of GHD n % (%) (%) (%) (%)

IdGHD (n ¼ 296) 38 13 82 34 48 5 CongGHD (n ¼ 142) 71 50 93 68 49 11 AcqGHD (n ¼ 245) 127 52 87 69 54 57

Table 3 Age and height at onset of GH therapy, birth weight and mid-parental height (SDS) in the three aetiology groups (means^S.D.).

Age at Height Birth onset at onset weight Mid-parental Group (years) (SDS) (SDS) height (SDS)

IdGHD 8.6^4.4 22.8^0.9 21.0^1.2 20.4^1.2 CongGHD 6.5^4.7 23.0^1.3 20.7^1.0 0.1^1.0 AcqGHD 10.9^3.2 21.4^1.1 20.3^1.0 0.2^0.9

Respectively, 90 and 86% of the patients with idGHD and congGHD were smaller than the 3rd percentile at the start of rhGH treatment, whereas only 35% of the patients with acqGHD were smaller than the 3rd percentile (P , 0.001). Age and height at onset of rhGH therapy did not change significantly during the studied period (data not shown). Birth weight was significantly below the population mean (P , 0.001) in the patients with idGHD and congGHD. Respectively, 18, 11 and 4% of the patients with idGHD, congGHD and acqGHD were considered as SGA (P , 0.001). Mid-parental height SDS of the patients with congGHD and acqGHD was comparable to the general population. In contrast, patients with idGHD had a mid-parental height below the normal population (P , 0.001); 8% of the patients in this latter group had a mid-parental height below the 3rd percentile while this was the case in only 3 and 1% of the patients with congGHD and acqGHD respectively. We did not find any major differences between boys and girls for the parameters detailed above except that in girls with congGHD and acqGHD, rhGH treatment was initiated at a younger age than in boys: 4.9 versus 7.2 years, and 10.2 versus 11.4 years (P ¼ 0.003) respectively. Figure 2 shows that 58% of the patients with congGHD started treatment before the age of 10 years; there was a bimodal distribution for the age at onset of rhGH therapy Figure 2 Age at onset of GH therapy in the three main groups: in children with idGHD, with a first peak around the age of (a) idGHD, (b) congGHD, (c) acqGHD. 4–5 years and another peak around the age of 10–12 years. In the patients with acqGHD, the age at onset of prevalence of childhood GHD in Belgium in 2001 to therapy ranged between 8 and 15 years. be 1/5600. This number is comparable to the obser- vations of the Dutch Growth Foundation (1/5000), Discussion using a comparable method to evaluate the prevalence of GHD (3). Our prevalence is somewhat lower than Based on data from children diagnosed and treated for that reported in the United Kingdom (1/4018) (2) GHD by the members of the BSGPE, we estimated the and in the state of Utah in the USA (1/3480) (1),

www.eje.org

Downloaded from Bioscientifica.com at 10/01/2021 07:58:02AM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2004) 151 Epidemiology of childhood GHD 71 while it is higher than in Australia (1/6757) (4), Japan hypoplasia, a missing, fine or normal pituitary stalk (1/6797) (5) and China (1/8646) (6). As pointed out and an ectopic posterior pituitary. In other studies by Lindsay and colleagues (1), the true prevalence of this ratio is not reported. GHD can only be derived from large-scale screening Although patients with congGHD have GHD from studies. In their survey evaluating 114 881 elementary birth, rhGH treatment was started at a mean age of school children, the presence of GHD was found in 16 6.5^4.7 years. The delay in starting rhGH therapy previously unrecognized children out of 555 with may be related to the degree of GH insufficiency associ- short stature (,3rd percentile) and poor growth rates ated with the various congenital conditions (19) and/or (,5 cm/year). Seventeen children were already the delay in diagnosis which depends on the moment of known to have GHD. Our prevalence number, based referral to the paediatric endocrinologist. In the idGHD on patients referred to paediatric endocrinologists, group a bimodal distribution for the age at onset of probably underestimates the true number of GHD chil- rhGH therapy is observed with peaks in mid and late dren, and a population-based screening is needed to childhood. It remains to be shown whether these two reveal the true prevalence of childhood GHD. cohorts correspond to different pathophysiological In our series of children with GHD, 41% were classified mechanisms. In contrast, the age range in the acqGHD as idGHD, 20% as congGHD and 35% as acqGHD. The group appears broader, from 8 to 15 years. These obser- distribution among the three main diagnostic groups of vations have also been reported by August et al. (15) and GHD has remained relatively constant over the past by Cutfield et al. for patients in the KIGS database (14). 16 years. This is in agreement with the results reported In the patients with idGHD, the age at onset of rhGH by Cutfield et al. for patients enrolled into KIGS (14). therapy (8.5 years) is lower than in the patients included However, the distribution amongst the various aetiolo- in KIGS (10.3 years) (18) or in France (11.1 years) (17) gies in our study differs from that reported in other while it is comparable to that of other countries such as countries or databases. The percentage of idGHD is the United Kingdom and Germany (8.2 years for the lower than that observed in other surveys where the 2 countries) (17). These differences can be due to vari- percentage of idGHD patients ranges from 65 to 85% ations in diagnostic procedures between countries (15–17). In contrast, in our series more patients were (e.g. sex steroid priming) and/or between the considered diagnosed as congGHD. These discordances can partly time periods in the evaluation of short children for GHD. be explained by classification differences: in less recent For the congGHD and acqGHD, the age at onset of rhGH studies, a number of patients with hypothalamo–pitu- therapy is comparable to other series (4, 15, 18). The itary malformations who have not been evaluated by a age at onset of rhGH therapy did not change signifi- cranial MRI may still have been classified in the idGHD cantly in Belgium between 1986 and 2001 for the group. Indeed, as the diagnostic procedures progress, three groups. This is in line with the data for the patients more abnormalities by imaging or at the gene level will with idGHD included in the KIGS database between be found in patients heretofore classified as idGHD. In 1987 and 1997 (16). addition, possible gene defects were not systematically In the patients with idGHD and congGHD rhGH treat- searched for in our idGHD patients. For these reasons, ment was started when a serious height deficit was the number of patients classified as idGHD will decrease already present. In patients with idGHD, height at in the future. Another possible difference between this onset of GH therapy (22.8 SDS) is comparable to that study and others is the greater proportion of peripubertal reported in KIGS (22.5 SDS) (18) but lower than in patients with constitutional delay of growth and puberty Sweden for example (22.3 SDS) (17). In patients and with low GH secretion included in the other surveys. with congGHD (23.0 SDS) and acqGHD (21.4 SDS) This latter hypothesis is supported by the finding that the mean height at onset is comparable to that reported age at onset of rhGH therapy in the idGHD group is lower in KIGS (18). In children with idGHD, Cutfield et al. in Belgium (8.5 years) compared with the patients reported a reduction in the height deficit at the start enrolled, for example, into KIGS (10.3 years) (18). of treatment in patients with idGHD and acqGHD of In line with the data in the literature, boys outnum- 0.4 and 0.6 SDS respectively over a period of 10 years bered girls in all aetiology groups (4, 15, 17, 18). In (1987–1996) (14). This was not the case in our study. patients with idGHD, the ratio of males to females In line with the data reported in KIGS (18), we in our study is, however, lower than that reported in observed that the mean birth weight was lower than other studies (1.5 in Belgium versus 2.5 (USA) (15), the population mean in the children with congGHD 3.1 (Sweden) (17), 1.9 (UK) (17), 2.0 (France) (17), (20.7 SDS) and idGHD (21.0 SDS). These data are 2.6 (Germany) (17) and 2.0 (KIGS) (18)). This is also in line with data suggesting that GH plays a role in pre- the case in patients with acqGHD (1.2 in Belgium natal growth (20, 21). In agreement with the data versus 2.1 (USA) (15) and 1.5 (KIGS) (18)). In contrast, from the literature (14, 18), mean mid-parental the ratio of males to females is higher in the Belgian height was below the population mean in the patients patients with congGHD (3.0 versus 1.3 (USA) (15) with idGHD, whereas parental height was normal in and 1.7 (KIGS) (18)) especially in patients with a cen- the other groups. This finding suggests that patients tral malformation that associates an anterior pituitary with familial short stature can present (transient)

www.eje.org

Downloaded from Bioscientifica.com at 10/01/2021 07:58:02AM via free access 72 M Thomas and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2004) 151

GHD (22), underlining the need for re-evaluation of GH 7 Parkin JM. Incidence of growth hormone deficiency. Archives of secretion in adulthood (9). Disease in Childhood 1974 49 904–905. 8 Audi L, Gilabert A, Lloveras G, Marti-Henneberg C, Rodriguez- In conclusion, we estimated the prevalence of child- Hierro F, Vilardell E et al. Long-term GH therapy: epidemiology hood GHD in Belgium to be 1/5600. Over the past 16 and auxologic outcome. Hormone Research 2002 57 113–119. years, the yearly number of new patients and the aux- 9 Thomas M, Massa G, Bourguignon JP, Craen M, De Schepper J, de ological parameters remained relatively constant. In Zegher F et al. Growth hormone (GH) secretion in patients with childhood-onset GH deficiency: retesting after one year of therapy comparison with the data from other surveys and and at final height. Hormone Research 2003 59 7–15. countries we observed a higher percentage of children 10 Ranke MB. The KIGS aetiology classification system. In Growth with congenital GHD. However, any comparison with Hormone Therapy in KIGS – 10 Years Experience, pp 389–401. other publications is difficult as diagnostic modalities Eds MB Ranke & P Wilton. JA Barth Verlag; Heidelberg, Leipzig, (e.g. MRI, genetic analyses, sex steroid priming) for 1999. 11 Tanner JM, Whitehouse RH & Takaishi M. Standards from birth to GHD not only vary between countries, but also between maturity for height, weight, height velocity and weight velocity: the considered time periods. Moreover, the occurrence British children 1965. Archives of Disease in Childhood 1966 41 of acqGHD in oncologic patients may be related to the 454–471, 613–635. tumour treatment which also varies between countries 12 Niklasson A, Ericson A, Fryer JG, Karlberg J, Lawtrence C & Karlberg P. An update of the Swedish references standards for and time periods. Nevertheless, our data underline the weight, length and head circumference at birth for given gesta- need for an accurate cranial MRI evaluation in all chil- tional age (1977–1981). Acta Paediatrica Scandinavica 1991 80 dren with GHD in order to detect those with central 756–762. malformations, and the need for genetic analysis in 13 Population et meńages. In Population Totale et Belge au 1.1.2001, those patients suspected of genetic causes of GHD. As pp 38. Ed Institut National de Statistique, Bruxelles, 2001. 14 Cutfield WS, Jo¨nsson P & Bagewitz A. Demographic trends: children with congGHD and idGHD are still diagnosed 1987–1996. In Biannual Report no. 15, pp 3–12. Stockholm: relatively late with a severe height deficit, growth of Pharmacia and Upjohn International Growth Database, 1997. all children, especially those with short stature, 15 August GP, Lippe BM, Blethen SL, Rosenfeld RG, Seelig SA, should be monitored very carefully, and timely referral Johanson AJ et al. Growth hormone treatment in the United States: demographic and diagnostic features of 2331 children. to a paediatric endocrinologist with experience in Journal of Pediatrics 1990 116 899–903. growth disorders should be considered. 16 Chatelain P. Trends in the diagnosis and treatment of short stature as revealed by KIGS. In Growth Hormone Therapy in KIGS – 10 Years Experience, pp 11–20. Eds MB Ranke & P Wilton. JA Acknowledgements Barth Verlag; Heidelberg, Leipzig, 1999. 17 Price DA, Johnston Betts PR, Buckler JM & Donaldson MD. Bio- We are grateful to the other members of the Belgian synthetic human growth hormone treatment in the UK: an Study Group for Paediatric Endocrinology who took audit of current practice. Archives of Disease in Childhood 1994 71 266–271. part in this work: V Beauloye, D Beckers, L Dooms, 18 Ranke MB. Lindberg A on behalf of the KIGS International Board. I François, M C Lebrethon, P Malvaux, R Rooman, Demographic characteristics of patients within KIGS: special S Tenoutasse, J Vanbesien, M Vandeweghe and M van emphasis on sex and onset of puberty. In Biannual Report no. Helvoirt. This work was supported by grants from the 18, pp 1–9. Stockholm: Pharmacia International Database, Foundation of the Belgian Study Group for Paediatric 2001. 19 Miyamoto J, Hasegawa Y, Ohnami N, Onigata K, Kinoshita E, Endocrinology. Nishi Yet al. Development of growth hormone and adrenocortico- tropic hormone deficiencies in patients with prenatal or perinatal onset hypothalamic hypopituitarism having invisible or thin pitu- References itary stalk on magnetic resonance imaging. Endocrine Journal 2001 48 355–362. 1 Lindsay R, Feldkamp M, Harris D, Robertson J & Rallison M. Utah 20 De Luca F, Bernasconi S, Blandino A, Cavallo L & Cisternino M. Growth Study: Growth standards and the prevalence of growth Auxological, clinical and neuroradiological findings in infants hormone deficiency. Journal of Pediatrics 1994 125 29–35. with early onset growth hormone deficiency. Acta Paediatrica 2 Vimpani GV,Vimpani AF, Lidgard GP,Cameron EH & Farquhar JW. 1995 84 561–565. Prevalence of severe growth hormone deficiency. British Medical 21 Gluckman PD, Gunn AJ, Wray A, Cutfield WS, Chatelain PG, Journal 1977 13 427–430. Guilbaud O et al. Congenital idiopathic growth hormone 3 Drent ML, Delemarre-Van de Waal HA & Wit JM. Eenmaal groei- deficiency associated with prenatal and early postnatal growth hormoon, altijd groeihormoon? De transitie van groeihormoonbe- failure. The International Board of the Kabi Pharmacia Inter- handeling van kind naar volwassenen. Nederlands Tijdschrift voor national Growth Study. Journal of Pediatrics 1992 121 920–923. Geneeskunde 2002 146 154–157. 22 Pozo J, Argente J, Barrios V, Gonzalez-Parra S, Munoz MT & 4 Werther G. Growth hormone measurements versus auxology in Hernandez H. Growth hormone secretion in children with treatment decisions: the Australian experience. Journal of Pedi- normal variants of short stature. Hormone Research 1994 41 atrics 1996 128 S47–S51. 185–192. 5 Tani N. Epidemiological study of dwarfism in Nigata prefecture. Folia Endocrinologica Japonica 1985 61 1295–1309. 6 Bao XL, Shi YF, Du YC, Liu R, Deng JY & Gao SM. Prevalence of growth hormone deficiency of children in Beijing. China Medical Received 10 December 2003 Journal 1992 105 401–405. Accepted 6 April 2004

www.eje.org

Downloaded from Bioscientifica.com at 10/01/2021 07:58:02AM via free access